Diane Ravitch’s blog, June 25, 2013
CREDO, the organization at Stanford University that analyzes charter performance, released the results of its latest national study.
Its earlier report (2009) showed disappointing results for charter schools, with only 17% outperforming traditional public schools. This was especially disappointing for the Walton Family Foundation, one of the main backers of CREDO.
The new study shows that charters are doing better than in 2009. They typically get about the same results as public schools, with some performing better, others performing worse.
I will do my own analysis later but meanwhile this is the best review I have seen, by Stephanie Simon of Reuters.
Key quote: “25 percent of charters outperformed nearby schools at teaching reading, while 19 percent did worse, and 56 percent were about the same. In math, 29 percent of charters did better, 31 percent did worse, and 40 percent were on par.”
The report raises many questions, implicitly, to a critical reader. Why is it that charter schools are not vastly outperforming public schools? They have the ability to skim and exclude. They have the benefit of “peer effects,” since they can expel troublesome students and send them back to their public school. Nearly 90% are non-union. They can fire teachers at any time and offer performance bonuses if they wish. They do everything that “reformers” dream of, yet they are hardly different in test scores overall from public schools, which typically must take all children and do not have the support of the Obama administration, major corporations, big media, big foundations, and hedge fund managers. The fact that charters serve large numbers of black, Hispanic, and poor students does not mean they serve a representative sample of students with disabilities and English language learners (they don’t). To compare a school that can select its student body with one that cannot is inherently unfair. The fact that the public schools do as well as the charter schools, despite their advantages, is remarkable.